TESTN Error estimate of discriminant for normal distribution.
E = TESTN(W,U,G,N)
DescriptionThis routine estimates as good as possible the classification error of Gaussian distributed problems with known means and covariances. N normally distributed data vectors with means, labels and prior probabilities defined by the dataset U (size [C,K]) and covariance matrices G (size [K,K,C]) are generated with the specified labels and are tested against the discriminant W. The fraction of incorrectly classified data vectors is returned. If W is a linear 2class discriminant and N is not specified, the error is computed analytically. See alsomappings, datasets, qdc, nbayesc, testc,
